In-vivo imaging studies of brain structures provide valuable information about the nature of neuropsychiatric disorders, including neurodegenerative diseases and/or disorders of abnormal neurodevelopment. Imaging can depict functional and morphologic information and has become an important component to detect normal biological variability and change from normal. Image acquisition shows steady progress with respect to spatial resolution, contrast-to-noise ratio, highspeed imaging, and versatility of scanning sequences measuring a variety of properties of tissue and function. To keep pace with advances in imaging and with needs of clinical research, image analysis research has to develop effective processing tools suitable for exploratory studies and for testing hypothesis in large clinical studies. Advanced image analysis methodology and statistical analysis methods, if developed on a sound mathematical foundation and on the principle of producing generic tools integrated into a common platform, will not only be applicable to human neuroimaging research as proposed here but will be relevant for imaging studies in cancer research, various other imaging domains like animal imaging studies, and even confocal microscopy imaging, for example.